Youhui ZHANG
Professor
Department of Computer Science and Technology
Joined Department: 2002
Email:zyh02@tsinghua.edu.cn
URL:https://craft.cs.tsinghua.edu.cn/
Phone:010-62785592
Fax:010-62771138
Education background
Bachelor of Computer Science & Technology, Tsinghua University, Beijing, China, 1998
Ph.D. in Computer Architecture, Tsinghua University, Beijing, China, 2002
Social service
Member of the Editorial Board of Computer Research and Development
Member of the Editorial Board of Journal of Computer Science and Technology
Areas of Research Interests/ Research Projects
Research Areas
High-Performance Processor Microarchitecture; Neuromorphic Computing Chips and Fundamental Software
Courses
An Introduction to Computer Systems (Undergraduate)
Teaching Profile
I primarily teach core undergraduate courses in computer systems. Together with other instructors in the systems course group (including those teaching Computer Organization, Operating Systems, Compilers, and teaching experiments), I actively participate in the "curriculum reform aimed at cultivating computer systems capabilities." This initiative has been recognized with the Second Prize of the National Teaching Achievement Award in Higher Education (2018) and the First Prize of the Beijing Municipal Teaching Achievement Award in Higher Education (2017).
Correspondingly, I have innovated the teaching content and structure of relevant courses. The undergraduate course I teach, "Assembly Language Programming," was designated as one of the first batch of National Quality Shared Courses (2016) and one of the first batch of National First-Class Undergraduate Courses (2020).
I engage in research-oriented education (where undergraduates play a leading role in processor design and tape-out) and practice-oriented education (actively developing professional practice courses). The Ph.D. students I have supervised have been awarded the title of Outstanding Ph.D. Graduate at Tsinghua University (2020) and the 2020 "CCF Outstanding Doctoral Dissertation Award" by the China Computer Federation (CCF).
Research Projects
The National High-Tech R&D Program of China (The 863 Program): Development of Test and Evaluation System of Cloud Computing (2013-2015);
The MOE-Intel IT Program: Research on the Transparent SaaS usage model of Existing Teaching Software (2011-2013);
The National High-Tech R&D Program of China (The 863 Program): Overall Technologies of High-end Fault-tolerance Computers (2008-2011);
The National Natural Science Foundation of China： Architecture Research of Storage Chip-Multiprocessors with Object-data Support and Dynamic Function-extension (2008-2010);
The National Basic Research Program of China (The 973 Program)：Research on the Virtualization of Resources from Multi-computers (2007-2011);
The National High-Tech R&D Program of China (The 863 Program): Research on the Portable and Personalized Software based on Virtualization and Network Services (2006-2008);
The National High-Tech R&D Program of China (The 863 Program): Middleware of High-performance Clusters (2002-2004);
The National High-Tech R&D Program of China (The 863 Program): Development of a 32-bit High-performance Embedded Processor (2002-2004).
Research Status
I conduct research on novel processor microarchitectures and the fundamental hardware / software of neuromorphic computing. Addressing the issues of ambiguous hierarchical interfaces and tight hardware-software coupling in neuromorphic computing systems due to the lack of a comprehensive theoretical foundation, we proposed the "neuromorphic computing completeness" inspired by the computational characteristics of neural networks, as well as a decoupled hardware-software hierarchy for neuromorphic computers and basic hardware primitives based on this completeness. It is demonstrated that such systems can support general-purpose computation, hardware completeness, and compilation feasibility for neuromorphic computing. These research findings have been applied to the development of the "TianjiC" neuromorphic chip, which targets artificial general intelligence, resolving the key issue of toolchain support for the fusion of heterogeneous networks. Based on these results, we designed a RISC-like neuromorphic chip and basic software that efficiently supports such hardware primitives, significantly enhancing performance. Subsequently, I proposed the design of a "general-purpose" neuromorphic computing chip and system, and leveraged the advantages of neuromorphic architecture to benefit general-purpose processors by designing a heterogeneous processor that integrates the strengths of neuromorphic architecture for general and sparse computing. This processor successfully underwent experimental tape-out in 2022 and 2025.
The paper on "neuromorphic computing completeness" was published in Nature (October 2020), marking the first Nature paper in the field of computer system architecture in China. It was recognized as a featured outcome at the 2020 World Internet Conference and one of the top ten global AI advancements by the BAAI (Beijing Academy of Artificial Intelligence). The "TianjiC" chip paper was published as a cover article in Nature (August 1, 2019), the first Nature paper in China's AI and chip fields, and was selected as one of China's top ten scientific advancements in 2019 and a leading scientific achievement at the 2019 World Internet Conference. The achievements have also received the Intel China Natural Science Achievement Award for Outstanding Research (2024), the First Prize for Technical Invention by the China Computer Federation (2022), and the Gold Medal at the Geneva Invention Awards (2022).
Honors And Awards
Intel China Natural Science Achievement Award for Outstanding Research (2024) ;
First Prize for Technical Invention by the China Computer Federation (CCF): Key Technologies and Applications of Heterogeneous Fusion Neuromorphic Computing Chips (2022);
Gold Medal at the Geneva Invention Awards: Heterogeneous Fusion Neuromorphic Computing Chip (2022);
2020 World Internet Conference Director's Special Recommended Achievement: A System Hierarchy for Neuromorphic Computing (2020);
2019 World Internet Conference Leading Scientific and Technological Achievement Award: The Heterogeneous Integrated Tianjic Chip for Artificial General Intelligence (2019);
2019 China's Top Ten Scientific Advances: Development of a Heterogeneous Chip for Artificial General Intelligence (2019);
National Teaching Achievement Award in Higher Education, Second Prize: Curriculum System Reform for Computer System Capability Cultivation (2018);
Ministry of Education, The People's Republic of China: Performance Analysis of High Performance Computers, Award for Science and Technology Progress, 1st Class (2009);
Chinese Institute of Electronics, Award for Science and Technology: Performance Analysis of High Performance Computers, Award for Science and Technology Progress, 1st Class (2009);
State Council, The People's Republic of China: High Performance Cluster Computers and Massive Storage Systems, National Award for Science and Technology Progress, 2nd Class (2007).
Academic Achievement
1. Youhui Zhang*; Peng Qu; Yu Ji; Weihao Zhang（the above are co-first authors）; Guangrong Gao; Guanrui Wang; Sen Song; Guoqi Li; Wenguang Chen; Weimin Zheng; Feng Chen; Jing Pei; Rong Zhao; Mingguo Zhao; Luping Shi*. A system hierarchy for brain-inspired computing, Nature, 2020, 586: 378-384.
2. Jing Pei; Lei Deng; Sen Song; Mingguo Zhao; Youhui Zhang; Shuang Wu; Guanrui Wang（the above are co-first authors）; Zhe Zou; Zhenzhi Wu; Wei He; Feng Chen; Ning Deng; Zheyu Yang; Cheng Ma; Guoqi Li; Wentao Han; Huanglong Li; Huaqiang Wu; Rong Zhao; Yuan Xie; Luping Shi*.Towards artificial general intelligence with hybrid Tianjic chip architecture, Nature, 2019, 572: 106-111.
3. Ji, Yu; Zhang, Youyang; Xie, Xinfeng; Li, Shuangchen; Wang, Peiqi; Hu, Xing; Zhang, Youhui*; Xie, Yuan; FPSA: A Full System Stack Solution for Reconfigurable ReRAM-based NN Accelerator Architecture, 24th ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS 2019), USA, 2019.
4. Xiaoyi Liu, Zhongzhu Pu, Peng Qu, Weimin Zheng, Youhui Zhang*. ActiveN: A Scalable and Flexibly-programmable Event-driven Neuromorphic Processor. In MICRO 2024.
5. Tingji Zhang, Boris Grot, Wenjian He, Yashuai Lv, Peng Qu*, Fang Su, Wenxin Wang, Guowei Zhang, Xuefeng Zhang, Youhui Zhang*. Hierarchical Prefetching: A Software-Hardware Instruction Prefetcher for Server Applications. In ASPLOS 2025.
